A postgraduate student has designed a device that gives robots the ability to smell. This technology could save lives by helping to locate the victims of natural disasters.
To initiate her research, Blanca Lorena Villareal studied the olfactory systems of living organisms.
Animals distinguish the source of an odour by registering the concentrations of the scent and the time elapsed as strength of the odour varies.
Villareal later applied mathematics to begin transforming the ideas into robotic realities.
Using artificial intelligence algorithms, she first developed a system that could recognise the smell of alcohol. Then, she altered the algorithms and developed them further to allow the detection of other scents.
This robotic olfactory system uses chemical sensors to function as nostrils. Data is then transmitted to a computer, where it is evaluated to determine the direction and proximity of the source of the smell.
“Unlike in other olfactory systems, this has the feature that in each cycle of ventilation the air chamber empties, making sensors ready for a new measurement,” explained Villareal, who developed the electronic nose as a postgraduate at the Monterrey Institute of Technology in Mexico.
Since the system can detect changes in the direction of the odour within one cycle, the robot can quickly identify and locate its source.
The device is compatible with various robotic platforms so it is not limited to any single application. As a result, robotic smell-tracking technology could prove useful in a number of fields.
Because the device can recognise odours such as blood, sweat and urine, it could track people trapped in dangerous situations as a result of natural disasters.
It is already being implemented into a project by the Mexican National Science and Technology Council to test its efficacy in emergency rescue situations.
Maybe one day the device could also be used to track intoxicated drivers and keep our roads safer, as well.
Villareal has been named as one of the most innovative young Mexicans by the Massachusetts Institute of Technology Review for her contributions to robotics, and is continuing her research by developing algorithms that will widen the variety of odours the robot can recognise.
She is also working to integrate smell-sensing into the robot’s decision-making process.
In the future, smell-tracking technology could even be programmed into androids to heighten their sensing capabilities and further humanise robots, even equipping them with a sharper sense of smell than the humans who created them.